PROJECT SUMMARY Ovarian cancer is among the deadliest cancer types, with a 5-year survival rate of only 47%. Survival rates for women with ovarian cancer have not changed in the past 25 years. This is partly due to the high frequency of patient relapses (over 75%) with cancers exhibiting drug resistance, making these cancers extremely difficult to treat effectively. Exacerbating the issue is that ovarian carcinomas are especially heterogeneous with respect to the cell of origin, genetics, and clinical evolution. These are major impediments to establishing effective experimental models for laboratory research, which are critical to improve the understanding and treatment of each patient's disease. Recent technological advances have enabled the development of `organoids' ? 3D self-organized tissue cultures ? from adult stem cells and subsequently from tumor samples. Tumor-derived organoids are arranged in a way that mimics the original tumor organization. Tumor-derived organoids that can be cultured long-term offer the advantage of extending the experimental lifetime of tumor resection samples, which are currently a limiting step in cancer research. Recent studies have shown that these organoids faithfully recapitulate the genetics, histology, and drug responses of original tumor samples from breast, colorectal, and pancreatic cancer patients, paving the way for `living biobanks' of these cancer types. Ovarian cancer organoids have been more challenging to establish, partly because of their heterogeneity and unique growth requirements ? but preliminary successes have now made it feasible to develop technology for ovarian cancer organoid derivation from fresh tumor samples. This project aims to build on these recent advances in the development of stem cell culture and stem-cell-based organoids to establish platform technology for ovarian cancer organoid generation. Numerous media conditions will be tested, which in a proof of concept will incorporate tumor genetic information to define specific growth requirements. The organoid models generated will be validated to ensure concordance with the original tumor genetics and histology, modifying the protocols accordingly. This project presents a unique opportunity to investigate whether organoid drug sensitivities correspond to patient treatment outcomes, as patients will be treated with standard-of-care chemotherapies following surgery to take the tumor samples, and organoids will be tested with the same agents. Finally, the protocols generated will be adapted for robotic automation so that numerous samples can be processed and biobanked in parallel at large scale, facilitating future adoption of these methods in a clinical setting. Taken together, this project will establish a new patient-specific preclinical model system to accelerate basic and clinical ovarian cancer research, ranging from disease mechanisms to personalized medicine approaches that will help to prioritize the treatments most likely to be effective for each patient.